INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

otine

-0.84

urrence

-0.82

etheless

-0.81

 Gamble

-0.81

caster

-0.79

otation

-0.79

eways

-0.76

ukong

-0.76

 warr

-0.76

imir

-0.72

POSITIVE LOGITS

âķĲ

0.65

bed

0.65

Ö¼

0.65

 knit

0.62

 closer

0.62

è¯

0.61

 Rated

0.61

 fence

0.59

 bites

0.59

 reprim

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.